Forums / The Science / Photometry / DSLR Photometry, another beginner's story
DSLR Photometry, another beginner's story
I have really enjoyed reading the comments in this forum about others experiences with DSLR photometry. I, too, have been experimenting with my DSLR, a Canon 20D. Starting last May I exchanged a number of emails with Des Loughney who was very helpful (and patient) with my "beginners" questions. Later, I joined the "Hopkins Campaign" and received more good advice from Jeff Hopkins. Like many of you, I am using AIP4WIN software. My biggest hurdle has been determining my camera’s Transformation Coefficient. I’ve read Des’ article and attempted the computations with a number of different star fields, but to no avail. My results always varied dramatically. Fortunately, AAVSO has recently added "tricolor-green" as one of the filter choices so I have been submitting my "un-transformed" observations using that selection. Also, Jeff suggested I try defocussing my images to eliminate saturated pixels. This has helped tremendously and now my "tricolor-green" results are coming within a few hundredths of a magnitude of others "V" band. Has anyone else been submitting observations using the "tricolor-green" filter selection? I’m hoping this is the correct choice for DSLR data that has not been converted to V band.
Hi Tom,That's interesting, are you saying that after the "defocussing-trick", your green-filter magnitude estimations match the V-band estimations of otrhers, so basically your Transformation Coefficient would be very close to zero, or was that referring to corrected magnitude values already? What is the TC you are using now?I too have struggled with the Transformation Coefficient (see the other beginners thread) and I'm now suspecting that the magic that AIP4Win does when reading in my Olympus raw images (applying camera white balance, de-bayerization with interpolation and what not...) is not helping at all to get good values. This might be different for the driver for the more popular Canon equipment, but when I load day-light "non-astronomical" images and load them into AIP4win with "apply camera white balance", they come out to be way too green. So i was thinking...if we are only interested in the green channel anyway in the end, why not try to work just with the untouched Green channel. Both AIP4win and IRIS (open source software with a similar feature set) allow you to read the raw Bayer-Matrix image. So workflow would be like this:-Read raw Bayer matrix-take just the green subpixels, zero out the other ones-dark frame calibration (where dark frames were processed likewise from raw Bayer matrix)-interpolate the green values for the missing (formerly red and blue) subpixels or maybe do a 2x2 binning of the green pixels to get a "normal" image with pixels for the green channel only.-stack-meassure differential magnitudes with AIP4win or IRIS-repeat and average results - file result under "Tri-color green" with AAVSOCUHeinz